https://store-images.s-microsoft.com/image/apps.32243.82b7f195-ec83-4631-b68c-2f40e3e566a9.550e0223-ccf2-47ce-9137-89b42c608fcc.473f82c6-5a3e-427b-a259-cffca43bbf7f

SQBM+ for Azure Learn & Development Plan

東芝デジタルソリューションズ株式会社 (本社)

カテゴリ
コンピューティングAI + 機械学習
サポート
サポート
法的情報
ライセンス契約プライバシー ポリシー

SQBM+ for Azure Learn & Development Plan

東芝デジタルソリューションズ株式会社 (本社)

A combinatorial optimization solver built on the quantum-inspired Simulated Bifurcation Algorithm.

SQBM+ is a set of solvers enabling users to quickly obtain nearly optimal solutions for large (e.g. millions of variables) combinatorial optimization problems, such as:
  • Portfolio optimization and risk management in financial service - 3D protein structure prediction in drug discovery process
  • Factory shift optimization in manufacturing or warehouse management
Key features:
  • Quickly obtains good approximate solutions for large optimization problems (QUBO, QPLIB and PUBO).
  • Suitable for QUBO problems with up to 10 million variables for the current version
  • Easy to use. Complicated parameter setting is not required.
SQBM is developed based on the theory described in the following papers:
  • Goto, H. et al. Combinatorial optimization by simulating adiabatic bifurcations in nonlinear Hamiltonian systems. Science Advances. 2019, 5, 4, eaav2372.
  • Goto, H. et al. High-performance combinatorial optimization based on classical mechanics. Science Advances. 2021, 7, 6, eabe7953.
詳細情報
User manual Product homepage Support policy Release notes